Arena-Hard (Arena-Hard-Auto-v0.1) Benchmark

Explore score history ↗

Arena-Hard (Arena-Hard-Auto-v0.1) is an AI benchmark in the Benchmark Radar catalog.

Released: 2024-06-18
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

Arena-Hard (Arena-Hard-Auto-v0.1) results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
gemini-2.5	—	79.0	—	evidence
gemini-2.5-flash	—	68.6	—	evidence
o3-2025-04-16	—	85.9	—	evidence
o4-mini-2025-04-16	—	74.6	—	evidence
o4-mini-2025-04-16-high	—	79.1	—	evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite Arena-Hard (Arena-Hard-Auto-v0.1)?

Arena-Hard (Arena-Hard-Auto-v0.1)